Help me with JFET problem!!!!

[RichJones]

ok... i am using the 2N3819 JFET with a 40v supply.

I am in need of the drain to sit at 20v. voltage gain around 6-10. around 2mA or so current.

currently I have a 10K drain resistor and a 1K source resistor.

The drain sits at 29v and my voltage swing has an assymetrical clipping.

Do I increase the drain resistor? If so decrease the source resistor????

The source resistor is NOT bypassed in anyway.

The drain is DC coupled to a MPSA18 connected as a common emitter with a 1K emitter resistor (want to run around 20mA).

cap coupled thereafter.

Thank you for any replies as I am in need of an answer asap and knew I could get help here.

[SY]

Disconnect the emitter follower and see what the unloaded FET stage does.

If doing this results in no change, I'd drop the source resistor a bit (increase of i_d) or substitute in another FET.

[RichJones]

ok...disconnecting the emitter follower changed the drain voltage by about 1v or so. decreasing the source resistor also increases circuit gain.

should i instead increase drain resistor higher than 10K????

hmmm........

what other JFET should I try??? different model # or just a new 2N3819?

txs!!!!

[Pedja]

RichJones,

I think you should lower your supply voltage and start to play safer. Check the datasheet.

Pedja

[SY]

So, what's the new drain voltage? What is the new gain? What does the clipping look like now?

Making the drain resistor larger will also increase circuit gain. 10K is probably a good value- let's keep that constant for the moment and attack the other variables.

[Jocko Homo]

Would be the easiest way. Yep, gain goes up.

You could make a cascode......the JFET may be happier with a lower V_ds. Ever think of a better JFET?

What exactly are you trying to build that needs 20 mA through a follower?


Jocko

[RichJones]

ok... it actually sits at 30v.

gain is around 6 with 10K / 1K combo of resistors.

changing to 10K with the 1K paralleled with 2K gain is 9 drain at 28v.

I didn't want to cascode them.... but I could try it I guess.

any other ideas???

and yes I know my supply voltage is a little high for these JFETS.

better JFET? like?

TXS!!!!

[SY]

OK, sounds like you're pretty close. Gain is right where you want it.

What kind of circuit are you trying to build?

[RichJones]

yeah gain is right but I need the drain to be at 20v.

so how do I do that???

a small preamplifier...

actually sounds really good.... but need to fix the assymetrical clipping issue.

[john curl]

You are exceeding the best working point of the 2N3819. RF fets, like the 3819, will leak excessively well below the max voltage point. This can cause noise, and make biasing difficult. A self cascode with the cascode device's gate connected to the source of the input fet, and the cascode source connected to the input fet's drain will work OK, but you should use a more modern fet, that one is almost 40 years old! I recommend a 2SK170 or a J113 (american type) fet for this voltage range.